Skip to content
Fan Zeng (Frank)
Back to resources

LibriTTS-R

LibriTTS-R applies learned audio restoration to the corresponding recordings from LibriTTS.

Visit official source (opens in a new tab)

Research use

Compare synthesis using original and restored recordings while controlling for speakers and text.

Limitations

Restoration can change acoustic characteristics or introduce artifacts. The release documents examples of unsuccessful restoration.

License and access

Dataset license
CC BY 4.0 (opens in a new tab)
Access status
Direct access

Use the publisher’s resource page and check the terms for the files and version you choose.

Additional sources